> ## Documentation Index
> Fetch the complete documentation index at: /docs/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Gateway

> A standalone routing layer for production LLM traffic. Reach 500+ models through one OpenAI-compatible endpoint with fallbacks and usage tracking.

AI Gateway is a standalone routing layer for production LLM traffic. Send requests through one OpenAI-compatible endpoint to access 500+ models across providers, with built-in fallbacks, retries, load balancing, and usage tracking.

## Quick Start

Complete the first request in a few minutes.

### 1. Create an account

[Sign up](https://router.toqan.ai/eu/auth/signup), then open the workspace and select **AI Gateway** in the sidebar. The workspace is your organization's account, its shared settings live under **Settings > Organization**, and it contains one or more projects that group resources and scope API keys.

### 2. Connect a provider

Use **Toqan Router** Credits, or add a provider key under **BYOK**. To add a key, open **BYOK**, choose [OpenAI](/docs/ai-studio/integrations/providers/openai.md), select Configure, and paste the key. See [Providers overview](/docs/ai-gateway/providers-overview.md).

### 3. Enable the model

Open the **Models** page (**Model List** in the **AI Gateway** sidebar), find `openai/gpt-5.4`, and toggle its **Status** to enable it. See [Using the router](/docs/ai-gateway/using-the-router.md).

### 4. Set up Toqan Router

Open **Settings > Organization > API Keys** and select Create API key. Pick the **Project** the key belongs to: requests made with the key are scoped to that project. Leave **Permissions** on **All** for this guide, because **Read only** or a **Restricted** key without **Responses** write access returns a 403 on the request below. See [API keys](/docs/ai-gateway/configuration/api-keys.md) for the other permission options. Create the key and copy it.

### 5. Run a request

Set the key in your terminal and send a request to the Responses endpoint. `$TOQAN_API_KEY` carries the key from the previous step as the bearer token.

```bash theme={"theme":{"light":"github-light","dark":"github-dark"}}
export TOQAN_API_KEY="your-api-key"

curl -X POST https://router.toqan.ai/eu/v3/router/responses \
  -H "Authorization: Bearer $TOQAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.4",
    "input": "Hello, world!"
  }'
```

For São Paulo, replace `/eu` with `/br`. See [Regions](/docs/toqan/regions.md).

### 6. Inspect the request

Open Traces to review the response, latency, token usage, and cost.

## Next steps

> [card] [Fallbacks & Retries](/docs/ai-gateway/features/retries.md)
>
> Keep requests flowing when a provider fails.

> [card] [Supported Models](/docs/ai-gateway/supported-models.md)
>
> Browse the model catalog and provider capabilities.

> [card] [Migrate to Toqan Router](/docs/ai-gateway/get-started/migrate.md)
>
> Move an existing application from the OpenAI SDK, OpenRouter, or LiteLLM.

> [card] [Model FAQ](/docs/ai-gateway/model-faq.md)
>
> Answers to common questions about models and their capabilities.

> [card] [Model Fusion](/docs/ai-gateway/model-fusion.md)
>
> Answer one request with a panel of models and a judge, under one model ID.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.